In operational testing programs with multiple cut scores, how does measurement precision, ability distribution, and cut-score location influence classification accuracy? Join us for a demonstration of the classification accuracy puzzle. Using data from an operational mastery test, we will demonstrate how classification accuracy is influenced by measurement precision at the cut score and cut score location relative to the distribution of ability. We will show how this interaction compounds when multiple cut scores are used because there is an opportunity for misclassification at each of the various cut score locations. We will offer suggestions and recommendations to improve classification accuracy. Our hope is that this information helps practitioners put the pieces together so they can set realistic expectations and design mastery tests that meet their classification accuracy goals.